The Active Sites of Fructose 6-Phosphate,2-Kinase

8/14/2019 The Active Sites of Fructose 6-Phosphate,2-Kinase

1/6

The Active Sites of Fructose 6-Phosphate,2-kinase:Fructose-2,6-bisphosphatase from Rat Testis

ROLES OF ASP-128, THR-52, THR-130, ASN-73, AND TYR-197*

(Received for publication, August 16, 1996, and in revised form, December 12, 1996)

Kosaku Uyeda, Xiao-Li Wang, Hiroyuki Mizuguchi, Yang Li, Cu Nguyen,and Charles A. Hasemann

From the Department of Veterans Affairs Medical Center, Research Service, Dallas, Texas 75216 and the Departments ofBiochemistry and Internal Medicine, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas 75235

To investigate the role in catalysis and/or substratebinding of the Walker motif residues of rat testis fruc-tose 6-phosphate,2-kinase:fructose-2,6-bisphosphatase(Fru 6-P,2-kinase:Fru-2,6-Pase), we have constructedand characterized mutant enzymes of Asp-128, Thr-52,Asn-73, Thr-130, and Tyr-197. Replacement of Asp-128 by Ala, Asn, and Ser resulted in a small decrease in Vmaxand a significant increase in Km values for both sub-

strates. These mutants exhibited similar pH activityprofiles as that of the wild type enzyme. Mutation ofThr-52 to Ala resulted in an enzyme with an infinitelyhigh Km for both substrates and an 800-fold decreasedVmax. Substitution of Asn-73 with Ala or Asp caused a100- and 600-fold increase, respectively in KFru 6-P withonly a small increase in KATP and small changes in Vmax.Mutation of Thr-130 caused small changes in the kineticproperties. Replacement of Tyr-197 with Ser resulted inan enzyme with severely decreased binding of Fru 6-Pwith 3-fold decreased Vmax. A fluorescent analog of ATP,2(3)-O-(N-methylanthraniloyl)ATP (mant-ATP) servedas a substrate with Km 0.64 M, and Vmax 25 milli-units/mg and was a competitive inhibitor with respect to ATP. When mant-ATP bound to the enzyme, fluores-

cence intensity at 440 nm increased. mant-ATP bindingof the wild type and the mutant enzymes were comparedusing the fluorometric method. The Kd values of theT52A and D128N enzymes were infinitely high and couldnot be measured, while those of the other mutant en-zymes increased slightly. These results provide evi-dence that those amino acids are involved in substratebinding, and they are consistent with the crystallo-graphic data. The results also suggest that Asp-128 doesnot serve as a nucleophile in catalysis, and since thereare no other potential nucleophiles in the active site, wehypothesize that the Fru 6-P,2-kinase reaction is medi-ated via a transition state stabilization mechanism.

Fru 2,6-P21

is the most potent activator of phosphofructoki-

nase (PFK), and its synthesis and degradation are catalyzed by

a bifunctional enzyme, Fru 6-P,2-kinase:Fru-2,6-bisphos-

phatase (Fru 6-P ATP i Fru 2,6-P2 ADP and Fru 2,6-P23 Fru 6-P Pi). Several isozymic forms of the enzyme from

mammalian tissues have been characterized (1). They are all

homodimers with Mr ranging from 108,000 to 120,000. The

primary structures of these enzymes revealed that the catalytic

domains are highly conserved, and the kinase and the phos-

phatase domains reside in the N-terminal half and the C-terminal half, respectively (27).

The reaction catalyzed by Fru 6-P,2-kinase follows ternary

complex formation (8) with direct transfer of the -phosphate of

ATP to the 2-OH of-D-Fru 6-P (9). Little is known about the

amino acid residues involved in substrate binding and cataly-

sis. The Fru 6-P binding sites of Fru 6-P,2-kinase have been

studied by chemical modification and site-directed mutagene-

sis. The result of affinity labeling experiments revealed that

Cys-107 and Cys-196 of rat liver Fru 6-P,2-kinase:Fru-2,6-Pase

and Cys-105 of the rat heart enzyme appear to be near or at the

Fru 6-P binding site (10, 11). Site-directed mutagenesis of the

rat testis enzyme demonstrated that Arg-102 (Arg-105 of the

liver enzyme) is essential for Fru 6-P binding (12). Similarly,

the importance of this Arg residue in the liver and muscleenzymes was shown by Rider et al. (13) and Kurland et al. (14)

by mutagenesis of the Arg residue to Ala, which resulted in a

200-fold increase in KFru 6-P. Arg-195 and Gly-48, respectively,

were shown to be essential for Fru 6-P and ATP binding in the

kinase reaction by site-directed mutagenesis in the liver

isozyme (15). More recently, Vertommen et al. (16) showed that

site-directed mutagenesis of Lys-54 and Thr-55 of the liver

isozyme (which correspond to Lys-51 and Thr-52 of the testis

enzyme) resulted in a 5000-fold decrease in the kinase activity.

Since mutation of Thr-55 to Cys resulted in loss of the kinase

activity but the mutated enzyme still binds mant-ATP and

ATP, they suggested that the Thr residue may be involved in

catalysis.

The Fru 6-P,2-kinase reaction is very similar to that cata-

lyzed by PFK (Fru 6-P ATP 3 fructose 1,6-bisphosphate ADP). The crystal structure of Escherichia coli PFK has been

solved (17). Based on the structures of complexes with Fru 6-P

and imidoadenosine 5-triphosphate (AMPPNP), Hellinga and

Evans (18) concluded that Asp-127 acts as a base, which in-

creases nucleophilicity by abstracting a H from the 1-OH of

Fru 6-P. In support of this idea is the observation that muta-

* This work was supported by a grant from the Department of Vet-erans Affairs, Grant DK16194 from the National Institutes of Health(to K. U.), and a grant from the Welch Foundation (to C. A. H.). Thecosts of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked adver-tisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

To whom correspondence should be addressed: Dept. of VeteransAffairs Medical Center, 4500 S. Lancaster Rd., Dallas, TX 75216. Tel.:214-372-7028; Fax: 214-372-9534.

1 The abbreviations used are: Fru-2,6-P2, fructose 2,6-bisphosphate;PFK, phosphofructokinase; Fru 6-P, fructose 6-phosphate; Fru-2,6-Pase, fructose-2,6-bisphosphatase; mant-ATP, 2(3)-O-(N-methylan-thraniloyl)-ATP; mant-ADP, 2(3)-O-(N-methylanthraniloyl)-ADP;

AMPPNP, 5-adenylylimidodiphosphate; WT, wild type; RT2K, fructose6-phosphate,2-kinase:fructose-2,6-bisphosphatase; NMP kinase, nucle-oside monophosphate kinase; HPLC, high pressure liquid chromatog-raphy; MES, 4-morpholineethanesulfonic acid; E

t, total enzyme

concentration.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 272, No. 12, Issue of March 21, pp. 78677872, 1997Printed in U.S.A.

This paper is available on line at http://www-jbc.stanford.edu/jbc/ 7867


2/6

tion of Asp-127 to Ser decreases the enzyme activity by 18,000-

fold compared with WT PFK (18). In the event that the mech-

anism of the Fru 6-P,2-kinase reaction is similar to that of

PFK, as has been widely presumed, then a nucleophile should

be identifiable that would activate the 2-OH of Fru 6-P for

attack on the ATP -phosphate.

Recently, we have crystallized (19) and solved the three-

dimensional structure of rat testis Fru 6-P,2-kinase:Fru-2,6-

Pase (RT2K) complexed with Mg2-ATP (20). These results

demonstrated that the Fru 6-P,2-kinase domain of the enzyme

does not resemble the structure of PFK as had been presumed

(21), but rather this domain is structurally similar to the family

of nucleoside monophosphate kinases (NMP kinases), includ-

ing adenylate kinase (22) and uridylate kinase (23). The Mg2-

ATP binding regions of the NMP kinases and the Fru 6-P,2-

kinase domain are remarkably similar, consisting of classical

Walker-A (GXXGXGKT) and -B (ZZZD; where Z represents a

hydrophobic amino acid) motifs (24), which comprise a phos-

phate binding loop and a Mg2-coordinating aspartate residue,

respectively. In the Fru 6-P,2-kinase domain, the conserved

Lys and Thr residues of the Walker-A motif are Lys-51 and

Thr-52, while Asp-128 is the conserved residue of the Walker-B

motif. No Fru 6-P was observed in the crystal structure, but the

Fru 6-P binding site can be accurately predicted both by ho-mology with the NMP binding sites of the NMP kinases and by

the location of residues previously identified to affect Fru 6-P

binding (1214). By modeling Fru 6-P in this presumed sub-

strate binding site, we have been able to search for a potential

nucleophile for the activation of the 2-OH of Fru 6-P. With the

exception of Asp-128, there is no Asp or Glu in the active site

that might fulfill this role. We have identified Asn-73, Thr-130,

and Tyr-197 as the only amino acids in the vicinity of the

binding site that might potentially act as weak nucleophiles

and/or assist in substrate binding.

To investigate the role in catalysis and/or substrate binding

of the conserved Walker motif residues and the weak nucleo-

philes in the substrate binding site, we have prepared mutant

enzymes at Asp-128, Thr-52, Asn-73, Thr-130, and Tyr-197.The kinetic properties and the nucleotide binding of these

mutant enzymes have been determined and are compared with

the WT enzyme.

EXPERIMENTAL PROCEDURES

Rabbit muscle PFK was prepared as described (25). The cDNA en-coding RT2K was prepared as described (7). Restriction enzymes, T4DNA ligase, and T4 polynucleotide kinase were purchased from New

England BioLabs (Beverly, MA). The Muta-Gene M13 in vitro mutagen-esis kit was purchased from Bio-Rad. The pT77 RNA polymerase/promoter plasmid (26) was a gift of Dr. Stan Tabor (Harvard MedicalSchool). The Sequenase version 2 sequencing kit was purchased fromU.S. Biochemical Corp. mant-ATP and mant-ADP were synthesized bythe method of Hiratsuka (27) and purified by HPLC on a Partisil 10SAX column (86 250 mm; Whatmann, Hillsboro, OR), eluting with 0.6M ammonium phosphate (pH 4.0) (28). The product was an equilibriummixture of 70% 3 isomer and 30% 2 isomer (28). All other chemicalswere reagent grade and obtained from commercial sources.

Site-directed MutagenesisPlasmid RT2K/pT77 containing the

RT2K gene cloned in a pT77 vector (7) was digested with XbaI andHindIII, and the isolated 1.7-kilobase pair fragment was ligated intothe XbaI-HindIII site of M13mp18 (M13/RT2K). The ligation mixturewas used to transform E. coli JM109 competent cells. The phage har-boring M13/RT2K was harvested and transfected into E. coli CJ236(dutung). The purified recombinant M13/RT2K phage was used toprepare uracil-containing single-stranded template. Synthetic oligonu-cleotide primers used for constructing various mutants are shown inTable I. The oligonucleotide-directed in vitro mutagenesis was per-formed as described by Kunkel (29) using the Muta-Gene M13 in vitromutagenesis kit. The double mutants designated as T52A/D128A and

T52S/D128A were constructed as follows. The pT77 vector containingD128A mutant of RT2K/pT77 DNA was digested with EcoRI, isolated,

and introduced into EcoRI-digested RT2K/pT77 containing the muta-

tion of T52A. The synthesized double-stranded DNA was used to trans-form E. coli MV1190 competent cells. Mutant derivatives were identi-

fied by DNA sequencing (30), and the DNAs were digested with NdeIand HindIII. The DNA fragments containing the mutated RT2K geneswere subcloned into the NdeI-HindIII sites of RT2K/pT77 and ex-pressed in E. coli as before (7). The WT and mutant Fru 6-P,2-kinase:

Fru-2,6-Pase enzymes were purified as described previously (31). How-ever, some of the mutant enzymes required slight modification of thisprocedure, which are described under Results.

Assay Method for Fru 6-P,2-kinaseThis assay was based on thedetermination of Fru 2,6-P2 and is the same as described previously (32)with slight modification. The reaction mixture in a final volume of 50 lcontained 100 mM Tris/HCl (pH 7.5), 0.1 mM EDTA, 10 mM MgCl2, 2 mM

ATP, and 2 mM Fru 6-P. The mixture was incubated at 30 C for 10 min. At the end of the reaction, 0.1 N NaOH (50 l) was added, and themixture was heated for 90 s at 80 C. Suitable aliquots were assayed forFru 2,6-P2 as described by Uyeda et al. (33). One unit of enzyme activity

is defined as the amount of enzyme that catalyzes the formation of 1mol of Fru 2,6-P2 /min under these conditions.

Assay Method for Fru-2,6-PaseThis assay measures the formationof Fru 6-P fluorometrically coupled to NADPH formation and wasdescribed previously (31). The reaction mixture (in a final volume of 1.0ml) contained 100 mM Tris/HCl (pH 7.5), 0.2 mM EDTA, 100 M NADP,0.4 unit of Glu 6-P dehydrogenase, 1 unit of phosphoglucose isomerase,and varying amounts of Fru 2,6-P2. The enzyme was desalted by columncentrifugation (34) in 15 mM Tris sulfate (pH 7.5), 0.5 mM EDTA, and 5mM dithiothreitol. The reaction was initiated with the addition of en-zyme and followed at 25 C by measuring the NADPH formation at

452-nm emission and excitation at 350 nm using an Aminco-BowmanSeries 2 luminescence spectrometer (SLM Aminco/Urbana, IL).

Fluorescence MeasurementsFluorescence spectra were determinedat 25 C with an SLM Aminco Bowman Series 2 spectrofluorometer.Excitation and emission slits were set to 4 nm. Spectra were correctedfor the buffer background but not for the instrument response func-tions. Binding of mant-ATP to enzymes was measured at 25 C in areaction mixture in a final volume of 0.2 ml, containing 50 m M Trisphosphate (pH 7.5), 0.1 mM EDTA, 5 mM dithiothreitol, and 1 M Fru6-P,2-kinase:Fru-2,6-Pase. Suitable aliquots (1 l) of 10 M mant-ATPwere added and corrected for dilution. Fluorescence intensity due to

free mant-ATP was measured in the absence of enzyme and used forcorrection. The excitation and emission were 280 nm (or 350 nm) and450 nm, respectively. The advantage of using 280-nm excitation wasthat thefluorescence of mant-ATP was much lower than that at 340 nm.

Separation and Determination of mant-ADPmant-ADP and mant-ATP were separated on a Partisil 10 SAX column (4.6 250 mm) usinga Dionex HPLC (Dionex/Sunnyvale, CA) and a Ratio-2 fluorometer(Optical Technology Devices, Elmsford, NY) equipped with a flow cell.mant-ADP was eluted from the column with a linear gradient of 0.250.5 M ammonium phosphate (pH 4.0), containing 25% ethanol with aflow rate of 1 ml/min. Under these conditions mant-ADP was usually

eluted after 7.5 min, while mant-ATP was eluted with 0.5 M ammoniumphosphate (after 15 min). To determine the concentration of mant-ADP,

a standard curve of mant-ADP was generated by the HPLC chromatog-

TABLE IOligonucleotides used for mutagenesis

These nucleotides are complementary to the DNA of RT2K at 182202 (T52A and T52S); 246267 (N73A and N73D); 410430 (D128AD128N); 419 436 (T130A); 416 436 (T130ST130V); 616 634 (Y197Fand Y197S).

Oligonucleotide Sequence

T52A 5-AGA AAT GTA GGC CTT GCC CCT-3T52S 5-AGA AAT GTA GGA CTT GCC CCT-3

N73A 5- TA C TG A CC C AC G GC G AA T TC C C- 3N73D 5- TA C TG A CC C AC G TC G AA T TC C C- 3

D128A 5-ATT GGT AGC AGC AAA AAC CGC-3D128S 5-ATT GGT AGC ACT AAA AAC CGC-3D128K 5-ATT GGT AGC CTT AAA AAC CGC-3D128E 5-ATT GGT AGC CTC AAA AAC CGC-3D128N 5-ATT GGT AGC ATT AAA AAC CGC-3

T130A 5-GGT GGT ATT GGC AGC ATC-3T130S 5-GGT GGT ATT GGA AGC ATC AAA-3T130V 5-GGT GGT ATT GAC AGC ATC AAA-3

Y197F 5-GTT TTC AAA GCA TTC AAT-3Y197S 5-GTT TTC AGA GCA TTC AAT-3

Active Site Residues of Fru 6-P,2-kinase7868


3/6

raphy of varying concentrations (010 pmol) of mant-ADP and by

calculating the concentration from the peak heights of the mant-ADPfluorescence peaks. Samples (10 l) containing mant-ADP were chro-matographed under the identical conditions, and the concentrations

were calculated using the standard curve.Other MethodsSDS-polyacrylamide slab gel electrophoresis was

performed with the Phast System (Pharmacia Biotech Inc.). Proteinconcentration was determined by the Bradford method (35) using bo-

vine serum albumin as a standard.

RESULTS

Expression and Purification of Various MutantsThe mu-

tated enzymes described here were purified using the same

procedure as that for the WT enzyme (31). However, D128N

and D128S mutant enzymes were eluted from the Yellow-3

column (fractionation step 4) with 20 mM ATP, while T52A and

Y197S were eluted from the same column with 0.2 or 0.3 M

potassium phosphate included in buffer A (Tris/P, 50 M, pH

7.5, 0.1 M EDTA; 0.1 M EGTA, 5% glycerol, 2 mM dithiothre-

itol, 1% polyethyleneglycol (M 300)). All of the enzymes were

purified to apparent homogeneity as judged by SDS-polyacryl-

amide gel electrophoresis.

Steady State Fru 6-P,2-kinase KineticsTable II summa-

rizes the kinetic parameters of the WT RT2K and various

mutant enzymes.

T52A and T52SThr-52 to Ala altered the Km for both Fru

6-P and ATP to infinite value, i.e. not saturable, and the Vmaxwas decreased to 11000 that of the WT enzyme. However, Thr-52

to Ser increased the Km and Vmax values about 2-fold. Similar

changes in the kinetic parameters were observed with T55C

and T55S mutants of rat liver Fru 6-P,2-kinase (16). Thus, the

side chain hydroxyl group appears to be essential for binding ofboth substrates and also for the catalysis. As depicted in Fig. 1,

the crystal structure showed that the side chain OH of Thr is

hydrogen-bonded to Mg2, which is chelated to the PO4 of

ATP (or 2-phosphate of Fru 2,6-P2; Ref. 20). Thus, this inter-

action may stabilize the pentacoordinated transition state. Ap-

parently, the extra methyl group present in Thr does not sig-

nificantly interfere with this interaction.

Asp-128 MutantsMutation of Asp-128 to Ala, Asn, and Ser

resulted in an 824-fold decrease in Vmax and a 30250-fold

increase and 33200-fold increase in KATP and KFru 6-P values,

respectively. The changes in the kinetic properties of the Ala

mutant were similar to those reported with the liver isozyme

(13). Asp-128 to Glu increased Vmax 2-fold, and Km values of

KATP and KFru 6-P increased 13- and 210-fold, respectively.

Surprisingly, the Asp to Lys mutation increased the Km values

comparable with the other mutants, but the Vmax was de-

creased by 1100. Since the carboxyl group of Asp is chelated to

Mg2 (Fig. 1), one would expect the negative charge to be

essential in the substrate binding and stabilization of the tran-

sition state. However, it appears that while the anionic inter-

action with Mg2 may not be essential for ATP binding, it is

required for catalysis.

T52A/D128A and T52S/D128ABoth of these double mu-

tant enzymes showed extremely high Km values for both sub-

strates as did the single mutant, T52A. The T52A/D128A mu-

tant displayed only trace kinase activity, but surprisingly, the

T52S/D128A mutant retained the same Fru 6-P,2-kinase activ-

ity as the WT enzyme although half of that of the single mutant

T52S. These results may suggest a more important role for

Thr-52 than for Asp-128 in efficient catalysis.Thr-130 MutantsRelatively small changes in the kinetic

properties, especially the Vmax values, were observed with Thr-

130 to Ala or Ser mutation. Mutation to Val increased KFru 6-P350-fold without affecting the Vmax. Thr-130 is located near the

furanose ring of Fru 6-P and may not have strong interaction

except to place the sugar moiety in proper orientation for effi-

cient catalysis.

Tyr-197 MutantsMutation of Tyr to Phe increased KATPand KFru 6-P by 6- and 70-fold, respectively, without change in

Vmax. However, mutation to Ser produced a large increase in

KFru 6-P

and a 19-fold increase in KATP. The Vmax decreased to13

that of the WT enzyme. These results suggest that Tyr-197

provides a hydrophobic pocket for Fru 6-P.

Asn-73 MutantsThe N73A mutant showed a 130-fold in-crease in KFru 6-P and a 4-fold increase in KATP, but the Vmax value was only slightly decreased (50% of WT). The N73D

mutant increased KFru 6-P to 600-fold and decreased Vmax to 9%

of WT without affecting KATP. The Vmax/KFru 6-PEt value of the

N73D mutant was 0.02% of WT, which was 121 that of N73A.

Asn-73 appears to be essential for Fru 6-P binding, probably

through hydrogen bonding, but not essential for ATP binding or

catalysis. This interpretation is consistent with the crystal

structure (Fig. 1), which shows that Asn-73 is located near the

furanose ring of Fru 6-P and may orient the Fru 6-P to a proper

position for catalysis through hydrogen bonding with the sugar

moiety. Introduction of a negative charge at Asn-73 (N73D)

may affect the interaction of Thr-52 and Asp-128, which are

essential for catalysis with Fru 6-P, and causes a large decrease

TABLE IIKinetic constants of rat testis Fru 6-P,2-kinase: Fru-2,6-Pase and the mutant enzymes

The results were averages of three determinations, and the standard errors were 15%. The range of substrate concentrations used for thedetermination of K

mvalues varied from 0.1 to 5 Km values.

EnzymeFru 6-P,2-kinase Fru-2,6-Pase

KATP KFru 6-P Vmax/Et V/KATP Et V/KFru 6-PEt KFru 2,6-P2 Vmax/Et V/KFru2,6P2Et

mM mM s1 M1 s1 M1 s1 M s1 M1 s1

WT 0.1 0.04 0.073 730 1800 0.04 0.013 330,000T52A 20 30 0.0001 0.08 0.036 450,000

T52S 0.19 0.11 0.15 790 1400 0.05 0.016 320,000D128A 2.9 5.0 0.0037 1.3 0.74 0.16 0.017 110,000D128N 25 1.3 0.0031 0.12 2.4 0.03 0.012 400,000D128S 7.5 8.0 0.0092 1.2 1.2 0.08 0.017 210,000D128K 2.2 4.5 0.00061 0.28 0.14 0.06 0.0079 130,000D128E 1.3 8.4 0.14 110 17 0.07 0.013 190,000T52A/D128A 20 30 0.0092 0.16 0.0092 58,000T52S/D128A 20 30 0.073 0.09 0.046 510,000T130A 0.16 0.28 0.070 440 250 0.06 0.027 450,000T130S 0.03 0.20 0.046 1500 230 0.07 0.044 630,000T130V 1.4 14 0.092 66 6.6 0.07 0.039 560,000N73A 0.4 5.3 0.037 93 7.0 0.12 0.028 230,000N73D 0.09 24 0.0082 91 0.34 0.04 0.011 280,000

Y197F 0.6 2.8 0.083 140 30 0.14 0.017 120,000 Y197S 1.9 30 0.023 12 0.13 0.031 240,000

Active Site Residues of Fru 6-P,2-kinase 7869


4/6

in Vmax/KFru 6-PEt.

Other Asp and Glu MutantsWe have altered all of the

highly conserved aspartates among the various isozymes to

Asn and some of the conserved glutamates to Gln includingAsp-94, -112, -160, -189, -205, -209, and -231 and Glu-97, -120,

-135, -155, and -195. These mutant enzymes were purified to

homogeneity, and their kinetic parameters were determined.

These mutations did not cause significant changes in the ki-

netic properties of the kinase activity compared with the WT

enzyme.

Fru-2,6-Pase Activities of the Mutant EnzymesIn general,

all of these mutant enzymes showed only small changes (24-

fold increase) in KFru 2,6-P2 and Vmax values of Fru-2,6-Pase.

mant-ATP as a SubstrateThe binding of mant-ATP, a flu-

orescent derivative of ATP, to various mutant enzymes and the

WT enzymes was compared. mant-ATP serves as a substrate

for the WT Fru 6-P,2-kinase, and the apparent Kmant-ATP and

Vmax

were 0.64 M and 25 milliunits/mg, respectively (data not

shown). It is surprising that the Km value for mant-ATP is

156-fold smaller than that for ATP, since subsequent examina-

tion of the crystal structure reveals that there is space avail-

able to accommodate the methylanthraniloyl substituent with

the potential for favorable packing interactions. To determine

if mant-ATP binds at the same catalytic site as ATP, Fru

6-P,2-kinase activity was measured at varying ATP or Fru 6-P

concentrations in the presence of constant concentration of 2

M mant-ATP. The kinase activity with mant-ATP alone as a

substrate in this reaction mixture (containing ATP as well) was

determined by measuring the formation of mant-ADP. The

mant-ADP and mant-ATP in the reaction mixture were first

separated by HPLC chromatography and quantitated fluoro-

metrically as described under Experimental Procedures. The

results demonstrated that mant-ATP was a competitive inhib-itor with respect to ATP but noncompetitive with respect to Fru

6-P. The KiATP and Ki

Fru 6-P values estimated from the plots

were 0.82 M and 2.8 M, respectively.

Binding of mant-ATP to WT Fru 6-P,2-kinase:Fru-2,6-Pase

was determined fluorometrically. When mant-ATP (2 M) was

bound to the enzyme, the fluorescence maximum decreased

from approximately 450 to 440 nm when excited at 350 nm, and

the fluorescence intensity increased compared with free mant-

ATP (Fig. 2, a versus c). The mant-ATP binding to the enzyme

required Mg2 (Fig. 2, b versus c), since there was no change in

the fluorescence in the absence of Mg2. The addition of 1 mM

ATP decreased the fluorescence at 440 nm, and the resulting

spectrum was identical to that of free mant-ATP (Fig. 2d),

suggesting that ATP displaced the enzyme-bound mant-ATP

completely. The same qualitative results were obtained when

mant-ATP was excited at 280 nm, but a much larger difference

in the fluorescent increase of bound mant-ATP was observed

compared with 350-nm excitation.

This increase in the fluorescence intensity was used to study

the binding of mant-ATP to the WT and some of the mutant

enzymes, and the dissociation constants calculated from the

double reciprocal plots are summarized in Table III. The dis-sociation constants for mant-ATP of the WT enzyme (0.80 M)

were comparable with the Kmant-ATP (0.64 M) and considerably

lower than KATP (100 M) and confirmed the tighter binding

than ATP. T52A and D128N did not bind mant-ATP, but other

mutant enzymes bound the ATP derivative as well as the WT

enzyme. Thus, the differences in the mant-ATP bindings to

these enzymes were similar to the differences in the KATP values (Table I), although the former values were nearly 2

orders of magnitude lower than the latter values (Table II).

These differences suggest that the binding of mant-ATP by the

enzyme was determined partly by hydrophobic interaction of

the mant- group.

These results, such as the large differences in the binding

constants of mant-ATP versus ATP, were in agreement with

FIG. 2. Emission spectra of mant-ATP bound to Fru 6-P,2-ki-

nase:Fru-2,6-Pase. The reaction mixture contained, in a final volumeof 0.2 ml, 50 mM Tris phosphate (pH 7.5), 0.1 mM EDTA, 5 mM dithio-threitol with 2 M mant-ATP (a), 1 M dimer (0.1 mg/ml) the rat testisenzyme, and 2 M mant-ATP (b), 1 M enzyme, 2 M mant-ATP, and 5mM MgCl2 (c); and 1 M enzyme, 2 M mant-ATP, 5 mM MgCl2, and 1mM ATP (d). The emission spectra were scanned at 25 C from 380 to560 nm with the excitation wavelength at 350 nm using an SLMspectrofluorometer Series 2.

FIG. 1. Stereo diagram of the sub-strate binding region of the Fru6-P,2-kinase domain of Fru 6-P,2-ki-nase:Fru-2,6-Pase. The side chains mu-tated in this study are drawn as ball andstick models, as is the Mg2-ATP. Thehydrogen bond interactions between the

ATP, Mg2, Thr-52, and Asp-128 are

drawn as solid white lines. Fru 6-P hasbeen modeled into the proposed Fru 6-Pbinding site based on 1) the coincidence ofan empty cavity in the crystal structurewith the substrate binding site of the ho-mologous NMP kinases and 2) the conver-gence of residues shown to affect the KFru6-P in this region. The proximity of Tyr-197, Thr-130, and Asn-73 to the ribosering of Fru 6-P is apparent.



5/6

those reported recently with the rat liver Fru 6-P,2-kinase and

its derivative containing six His residues (H6) at the C termi-

nus (16). However, there were several important differences.

The rat liver enzyme (H6) contains high and low affinity bind-

ing sites for Mg-mant-ATP and mant-ATP with Kd values of

108 M and 105 M, respectively. Moreover, mant-ATP binding

to the low affinity site is unaffected by mutation of Thr-55 and

Lys-54. We found that the rat testis enzyme had one binding

site for Mg-mant-ATP with the Kd values ranging from 107 to

106 M, and as shown in Fig. 2, its binding requires Mg 2. Themutant enzymes showed different Kd values, which were con-

sistent with the changes in the KATP values. These differences

may be due to the differences between these isozymes, which

may be a reflection of differences in the regulatory domains

since their catalytic domains are very similar.

pH Activity ProfilesIf Asp-128 is the nucleophile involved

in catalysis, mutation of this residue may show an altered pH

activity profile. To investigate this possibility, the Fru 6-P,2-

kinase activity of the WT, D128A, and D128N enzymes at pH

values between 6 and 9.5 were determined. The results shown

in Fig. 3 indicated that the WT enzyme showed the pH opti-

mum between 7 and 8.5, and the apparent pK values for the

ascending and the descending rims were 6.6 and 9, respec-

tively. The pH profiles of the mutant enzymes were similar tothat of the WT enzyme. The results suggested that Asp-128

does not serve as a base in the catalysis, confirming the muta-

tion results. These pH activity curves are completely different

from that of rat muscle bifunctional enzyme in which the ki-

nase activity increases linearly from 6 to 10, and the optimum

is pH 10 (13). Furthermore, mutation of Asp-128 to Ala of the

muscle enzymes results in loss of the pH optimum at 10 and

becomes independent of pH. Since the active sites of all of these

isozymes are highly conserved, it is unlikely that these differ-

ences are due to tissue-specific isozymic differences.

DISCUSSION

All kinase reactions or phosphoryl transfer reactions require

a base (nucleophile) for catalysis, positive charges to neutral-

ize/stabilize the negative charges of the ATP phosphates, andresidues to position the substrates and stabilize the transition

state. The determination of the three-dimensional structure

has demonstrated that the active site of the Fru 6-P,2-kinase

domain of the Fru 6-P,2-kinase:Fru 2,6-P2 appears to satisfy

these requirements with the exception of a catalytic nucleo-

phile. It was the goal of this study to probe by mutagenesis the

residues in the active site for their potential role as the appar-

ently missing nucleophile. The effect of mutagenesis of the

residues located at or near the active site presented herein is

consistent with the three-dimensional model determined by

crystallography as shown in Fig. 1. Thr-52 and Asp-128 are

hydrogen-bonded/salt-bridged to the Mg2 ion of the Mg2-

ATP complex. Accordingly, all but the most conservative mu-

tations of these residues have a very significant impact on the

KATP. Thr-52 is also close enough to the -phosphate of the ATP

to be directly hydrogen-bonded to it, and this may explain the

most extreme effect on KATP of the T52A mutation. The pro-

found effect on Vmax of the T52A mutation and the comparably

modest effects of the Asp-128 mutations were not predicted. We

would have expected the charge interaction of Asp-128 with the

Mg2 ion to dominate the coordination of the Mg2-ATP, while

our results indicate that Thr-52 is the more important Mg2

ligand. The Mg2 ion may play a role in catalysis by stabilizing

the negative charge involved in transition state formation.

Thr-52 may have an additional role in the ATP binding. Ac-

cording to the crystal structure of the bifunctional enzyme, the

kinase active site is nearly identical to the GTP binding site of

p21ras, and Thr-52 in the former is situated at the position

analogous to Ser-17 in the latter (36, 37). Muegge et al. (38)

proposed that the interaction of Ser-17 with the GTP -phos-

phate maintains a conformation of the P-loop main chain at-

oms, which optimize the orientation of the main chain dipoles

for interaction with the GTP phosphates. Thus, a component of

the decreased catalysis observed in our T52A mutants might be

due to a loss of an essential main chain conformation in the

P-loop, which promotes optimal electrostatic interactions with

the ATP. Further mutagenesis and crystallography of mutant

enzymes or transition state mimics will be necessary to resolve

the structural basis for this observation.

It should be noted that the observed changes in the kineticproperties of the mutated enzymes were not due to global

structural changes for the following reasons. 1) All of these

enzymes were purified to homogeneity with the same proce-

dure as that used for the WT enzyme. The only exception was

that some of the mutants bound to an affinity column (Yel-

low-3) more tightly than the WT enzyme, requiring slightly

higher ATP concentration and/or higher phosphate for elution,

although they showed increased Km values for the substrate(s).

2) Intrinsic tryptophan fluorescence spectra of Asp-128 and

Thr-52 mutants were identical to that of the WT enzyme (data

not shown).

The effect on the KFru 6-P of the mutants, which would ap-

parently only affect Mg2-ATP binding, can best be explained

by assuming that ligand binding in the Fru 6-P,2-kinase active

TABLE IIIDissociation constants of mant-ATP

The Kd

values were the average S.E. of three determinations. T52Aand D128N did not bind mant-ATP. The binding was determined flu-orometrically as described under Experimental Procedures.

Enzyme Kdmant-ATP

M

WT 0.80 0.05T130A 0.90 0.10

R136L 1.9 0.10T52S 1.6 0.10T52AD128N

Y197F 1.5 0.10

FIG. 3. Fru 6-P,2-kinase pH activity profiles of WT, D128A, andD128N Fru 6-P,2-kinase:Fru-2,6-Pase. Fru 6-P,2-kinase activity wasdetermined as described under Experimental Procedures, except thebuffer mixture consisted of 0.1 M MES, 0.051 M N-ethylmorpholine, and0.051 M diethanolamine at varying pH values, and 0.1 and 20 m M Fru

6-P for the WT and the neutral enzymes, respectively, were used. Thereaction was initiated with 10 mMATPand 20m M MgCl2 forthe mutantenzymes and with 1 mM ATP and 20 mM MgCl2 for the WT enzymes.

Active Site Residues of Fru 6-P,2-kinase 7871


6/6

site is an ordered process in which ATP binds to the enzyme

first followed by Fru 6-P. Earlier kinetic studies (8) suggested

that the addition of the substrates to liver Fru 6-P,2-kinase is

random, but it is possible that under these assay conditions the

ordered substrate addition was favored. Thus, mutations af-

fecting the ability to bind ATP will concomitantly affect Fru 6-P

binding, while the inverse is not necessarily true.

While the crystal structure of the Fru 6-P,2-kinase:Fru-2,6-

Pase did not include Fru 6-P bound at its active site, the

homology of this enzyme with the NMP-kinases and the loca-

tion of several residues known to affect KFru 6-P have allowed

for a confident prediction of the Fru 6-P binding site. As shown

in Fig. 1, there are several residues in the vicinity of the Fru

6-P binding site of the enzyme that are potential (albeit weak)

nucleophiles, including Thr-130, Tyr-197, and Asn-73. While

mutations of each of these residues (or Thr-52 and Asp-128) has

an impact on KFru 6-P (or KATP), their low impact on Vmax(1150-fold reductions) would indicate that none of them act as

a nucleophile for activating substrate (compared with the

13,000-fold reduction observed for the D127A mutation of the

PFK nucleophile (39)). Again, turning to the structure in Fig. 1,

this is reasonable, since their relation to a modeled Fru 6-P

substrate places them in positions to bind the ribose moiety

(Asn-73, Thr-130, Tyr-197) or the Mg2

-ATP (Thr-52, Asp-128)and not in the immediate vicinity of the gap between the

-phosphate and the Fru 6-P 2-OH, where an activating nu-

cleophile would be expected to reside. In addition, we have

mutated all of the highly conserved Asp and Glu residues in the

Fru 6-P,2-kinase domain, but none of these mutant enzymes

were affected significantly.

The evidence presented here supports our contention that

the catalytic mechanism of the Fru 6-P,2-kinase reaction may

not involve a strong nucleophile but instead is mediated via

transition state stabilization. This may be the reason for the

extremely low value for kcat of 0.07 s1 of Fru 6-P,2-kinase.

This is in contradiction to the widely held belief that the Fru

6-P,2-kinase reaction would be PFK-like (21). This is not

surprising in light of the structural differences between the Fru6-P,2-kinase domain and the PFK structures and the unantic-

ipated similarity between the Fru 6-P,2-kinase domain and the

NMP-kinase and G-protein structures. These latter enzymes

are believed to catalyze their reactions by a transition state

stabilization mechanism.

AcknowledgmentsWe thank Drs. Paul A. Srere and Paul F. Cookfor critical reading of this paper.

REFERENCES

1. Uyeda, K. (1991) in CRC Review (Kuby, S. A., ed) Vol. 2, pp. 445456, CRCPress, Boston, MA

2. Rider, M. H., Marchand, M. J., Hue, L., and Rousseau, G. G. (1987) FEBS Lett.

2243, 3173213. Algaier, J., and Uyeda, K. (1988) Biochem. Biophys. Res. Commun. 153,

3283334. Lively, M. O., El-Maghrabi, M. R., Pilkis, J., DAngelo, G., Colosia, A. D.,

Ciavola, J.-A., Fraser, B. A., and Pilkis, S. J. (1988) J. Biol. Chem. 263,839849

5. Crepin, T. M., Darville, M. I., Michel, A., Hue, L., and Rousseau, G. G. (1989)Biochem. J. 264, 151160

6. Sakata, J., and Uyeda, K. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 495149557. Sakata, J., Abe, Y., and Uyeda, K. (1991) J. Biol. Chem. 266, 1576415770

8. Kitajima, S., Sakakibara, R., and Uyeda, K. (1984) J. Biol. Chem. 259,

689669039. Kountz, P. D., Freeman, S., Cook, A. G., El-Maghrabi, M. R., Knowles, J. R.,

and Pilkis, S. J. (1988) J. Biol. Chem. 263, 160691607210. Sakakibara, R., Kitajima, S., Hartman, F. C., and Uyeda, K. (1984) J. Biol.

Chem. 259, 140231402811. Kitamura, K., Uyeda, K., Hartman, F. C., Kangawa, K., and Matsuo, H. (1989)

J. Biol. Chem. 264, 6344634812. Tsujikawa, T., Watanabe, F., and Uyeda, K. (1995) Biochemistry 34,

63896393

13. Rider, M. H., Crepin, K. M., deChoedt, M., Bertraud, L., Vertommen, D., andHue, L. (1995) Biochem. J. 309, 341346

14. Kurland, I., Chapman, B., Lee, Y-H., and Pilkis, S. J. (1995) Biochem. Biophys.Res. Commun. 213, 663672

15. Li, L., Lin, K., Kurland, I. J., Correia, J. J., and Pilkis, S. J. (1992) J. Biol.Chem. 267, 43864393

16. Vertommen, D., Bestrand, L., Sontag, B., DiPietro, A., Louckx, M. P., Vidal, H.,

Hue, L., and Rider, M. H. (1996) J. Biol. Chem. 271, 178751788017. Shirakihara, Y., and Evans, P. R. (1988) J. Mol. Biol. 204, 973994

18. Hellinga, H. W., and Evans, P. R. (1987) Nature 327, 437439

19. Istvan, E. S., Hasemann, C. A., Kurumbail, R. G., Uyeda, K., and Deisenhofer,J. (1995) Protein Sci. 4, 24392441

20. Hasemann, C. A., Istvan, E. S., Uyeda, K., and Deisenhofer, J. (1996)Structure4, 10171029

21. Bazan, J. F., Fletterick, R. J., and Pilkis, S. J. (1989) Proc. Natl. Acad. Sci.U. S. A. 86, 96429646

22. Muller, C. W., and Schulz, G. E. (1992) J. Mol. Biol. 224, 15917723. Muller-Dieckmann, H-J., and Schulz, G. E. ( 1995) J. Mol. Biol. 246, 522530

24. Walker, J. E., Saraste, M., Runswicks, M. J., and Gay, N. J. (1982) EMBO J.1, 945951

25. Uyeda, K., Miyatake, A., Luby, L. J., and Richards, E. G. (1978) J. Biol. Chem.253, 83198327

26. Tabor, S., and Richardson, C. (1985) Proc. Natl. Acad. Sci. U. S. A. 82,

1074107827. Hiratsuka, T. (1983) Biochim. Biophys. Acta 742, 496508

28. Hazlett, T. L., Moore, K. J. M., Lowe, P. N., Jameson, D. M., and Eccleston, J.F. (1993) Biochemistry 32, 1357513583

29. Kunkel, T. A. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 48849230. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl.Acad. Sci.U. S. A.

74, 5463546731. Tominaga, N.,Minami, Y.,Sakakibara, R.,and Uyeda,K. (1993)J. Biol.Chem.268, 1595115957

32. Furuya, E., and Uyeda, K. (1981) J. Biol. Chem. 256, 7109711233. Uyeda, K., Furuya, E., and Luby, L. J. (1981) J. Biol. Chem. 256, 83948399

34. Penefsky, H. S. (1977) J. Biol. Chem. 252, 2891289935. Bradford, M. M. (1976) Anal. Biochem. 72, 248254

36. DeVos, A. M., Tory, L., Milburn, M., Matias, P. M., Jancarik, J., Noguchi, S.,Nishimura, K., Miura, K., Ohtsuka, E., and Kim, S. H. (1988) Science 239,888893

37. Pai, E. F., Kabsch, W., Krengel, U., Holmes, K., John, J., and Wittinghofer, A.(1989) Nature 341, 209214

38. Muegge, I., Schweins, T., Langen, R., and Warshel, A. (1996) Structure 4,475489

39. Berger, S. A., and Evans, P. R. (1992) Biochemistry 31, 92379242


Documents

The Active Sites of Fructose 6-Phosphate,2-Kinase